C57BL/6, B6.129P2(C)-Cx3cr1^{tm2.1(cre/ERT2)Jung} /J (Cx3cr1^CreER) and B6.Cg-Gt(ROSA)26Sor^{tm14(CAG-tdTomato)Hze} /J (tdTomato) were purchased from the Jackson Laboratory. Male Cx3cr1^CreER were crossed with tdTomato mice to generate a Cx3cr1^CreER/tdTomato lineage reporter. To induce the lineage labeling during the embryonic stage, pregnant females were treated by oral gavage with either 2 mg/mL tamoxifen (Sigma-Aldrich) plus 0.7mg/mL progesterone (Sigma-Aldrich); 3 mg/mL tamoxifen plus 1.4 mg/mL progesterone; or 4 mg/mL tamoxifen plus 2.1 mg/mL progesterone at specified embryonic stage (E7.5 vs. E13.5). Due to the impact of tamoxifen on the pregnancy, progesterone was added to aid development and prevent spontaneous abortion. In addition, caesarian section and fostering were performed to allow growth to adulthood. As previously described, bone marrow chimeras were generated using busulfan (11). Experiments were conducted per National Institute of Health guidelines and protocols approved by the Animal Care and Use Committee at Duke University.

LPS (12.5 µg in 50 µL of PBS, Sigma-Aldrich) or PBS was administered by intranasal instillation (i.n.) under light isoflurane anesthesia. This occurred every other week for 3 doses. Following LPS or PBS exposure, the mice were allowed to recover for >8 weeks before acute HDM exposure. For the acute model, mice were challenged with 50µl of Dermatophagoides Pteronyssinus HDM (100µg total protein, Greer Laboratories Lot#296771, Lenoir, NC) or PBS intranasally (i.n.) on day 1, 7, and 14 under light isoflurane anesthesia. HDM Lot#296771 contains Der P1 113.21 µg/vial and endotoxin 28,750 EU/vial. Airway physiology measurements, bronchoalveolar lavage, and tissue harvests were conducted 48 hours after the last challenge.

Airway responsiveness to acetyl-β-methacholine (MCh; Sigma-Aldrich) was measured 48h following the final HDM exposure using a computer-controlled small animal ventilator (FlexiVent FX, Scireq, Montreal, Canada) as we have previously described (12). Mice were anesthetized with urethane (1–2 g/kg; Sigma-Aldrich), tracheotomized and intubated with a metal endotracheal tube (18G blunt needle), placed on a 37°C water-heated pad, and mechanically ventilated at a rate of 150 breaths/min with a tidal volume of 10 ml/kg and a positive end-expiratory pressure (PEEP) of 3 cm H₂O. To block spontaneous respiratory effort, mice were given pancuronium bromide i.p. (0.8 mg/kg; Sigma-Aldrich) five minutes before assessing airway responses. Both 1-second single-frequency and 3-second multi-frequency forced oscillation waveforms were applied using the Flexiware 7.6 software default mouse inhaled dose-response script to measure respiratory mechanics. The resulting pressure, volume, and flow signals were used to determine total respiratory system resistance (R_rs) and elastance (E_rs) or Newtonian resistance (Rn), tissue damping (G), and tissue elastance (H) as previously described (13, 14). MCh (10–100 mg/ml) was aerosolized using the FlexiVent FX aeroneb nebulizer attachment. The peak response at each dose was averaged and graphed along with each group’s average peak baseline measurement.

As previously described, the BALF was obtained using the gravity fill method (12). The lungs are inflated to 20 cm H20, approximating total lung capacity with PBS+100µM EDTA+40 µM DTPA via gravity inflation three times, and the fluid is withdrawn with a syringe inserted into a side port of the tracheal tubing. BALF was spun to remove cells. The mean BALF volume was 2.936 mL (standard deviation +/- 0.1984). The BALF was concentrated to 0.25mL using a centrifugal filter (Millipore), and cytokines were measured using a custom Procartaplex (Invitrogen). Cell pellets were treated with RBC lysis buffer (BioLegend) and re-suspended in 1mL HyClone (Thermo Scientific). Total cells were counted with a K2 Cellometer (Nexcelom Bioscience) using AO/PI staining to exclude dead cells/debris. After staining with Hema 3 solution (Protocol, Kalamazoo, MI), counts of different immune cells were obtained under light microscopy. The cell count is normalized to the initial volume of BALF (cells/ml). Cell counts, differentials and cytokines were assessed by technician blinded to the exposure conditions.

Flow cytometry was performed on the whole lung tissue using our previously published protocols (15, 16). In brief, mice were administered 50uL of 1000/mL heparin subcutaneously before euthanasia. Ten minutes after the injection, the mice were euthanized with isoflurane. The chest wall was opened, and the trachea cannulated. Following cardiac perfusion with 10mL of PBS to remove intravascular red cells, the lung was insufflated past total lung capacity with a digestion solution (1.5mg/ml of Collagenase A (Roche) and 0.4mg/ml DNase I (Roche) in HBSS plus 5% fetal bovine serum and 10mM HEPES). The lung tissue was excised and incubated at 37°C for 30 minutes with intermittent vortexing to generate a cell suspension. After digestion, the samples were washed and strained through a 70uM cell strainer and underwent red cell lysis (ACK RBC lysis solution). The cells were then manually counted and underwent staining for flow cytometry using a previously described flow cytometry panel and gating strategy [ Supplementary Table 1 , Supplementary Figure S1 , and (15)] of established markers of lung immune cells (17). Flow cytometry was performed on a BD LSRII using BD FACSDiva software and then analyzed using FlowJo version 10 software.

Left lung lobes were isolated and fixed by gravity inflation (20 cm H₂O) and immersion in 10% formalin. Lungs were paraffin-embedded, cut into 5µm sections, and stained by AML laboratories (Saint Augustine, FL). After individual staining, 10 images of randomly chosen variable-size airways were photographed at 10–20x magnification and analyzed in a blinded examination. Lung tissue inflammation was semi-quantitatively determined from hematoxylin and eosin (H&E) and Alcian blue-Periodic acid-Schiff (AB-PAS) stained sections using a four-tiered scoring system as previously described (18, 19). In addition, peri-bronchial fibrosis was determined from Masson’s trichrome staining using Image J (NIH) to calculate the percentage of peri-bronchial trichrome staining within the total tissue area containing the airway, including the airway’s lumen. Ten airways per animal were analyzed.

Representative unstained formalin-fixed paraffin-embedded (FFPE) sections (5 μm thickness) were stained using automated immunofluorescent techniques on BOND™ RXm Processing Module, utilizing the Bond Research Detection kit (Leica Microsystems) and Opal fluorophores (Akoya Biosciences) according to manufacturer instructions. In brief, FFPE sections were deparaffinized, hydrated with alcohol, and subjected to heat-induced epitope retrieval with Bond ER1 program. Slides were washed with Bond wash solution, and non-specific protein binding was blocked with protein block (5 min). For CD206 and Cathepsin K (CTSK) co-staining, the sections were sequentially incubated with CD206 (E6T5)(0.04ug/ml; Cell Signaling Technology) (30 min), Rabbit Polymer (10 min), Opal 520 (10 min) and CTSK (Rabbit anti-mouse CTSK polyclonal antibody(1:100; Novus Biological) (30 min), Rabbit Polymer (10 min), Opal 570 (10 min) Ab fluorophore combination. The nuclei were stained with spectral DAPI (4,6-diamidino-2-phenylindole) solution (Akoya Biosciences). Finally, the sections were cover-slipped using Vectashield HardSet Antifade mounting media (Vector Laboratories). Images were acquired using a Zeiss Upright 780 confocal microscope and processed using FUJI/Image J software.

Yolk sac macrophages (CD45⁺CD11b⁺F4/80⁺) were sorted from the yolk sac layer of E9.5 embryos. AMØs (CD45⁺CD64⁺SSC^hiCD11b^-CD11c⁺SiglecF⁺), IMØs (CD45⁺CD64⁺SSC^hiCD11b⁺CD11c^-SiglecF^-), Ly6C⁺ classical monocytes (CD45⁺CD64^loSSC^loCD11b⁺CD11c^-Ly6C⁺), or Ly6C- non-classical monocytes (CD45⁺CD64^loSSC^loCD11b⁺CD11c^loLy6C^-) were sorted from lungs of mice. 10,000 cells were lysed using SuperAmp™ lysis buffer according to manufacturer instructions. Agilent Whole Mouse Genome Oligo Microarrays 8x60K were used to analyze transcriptomic profiles of sorted cells. Microarray processing, including RNA extraction, cDNA synthesis and labeling, hybridization, imaging, and data analysis with normalization, were performed by Miltenyi Biotec. Subsequent hierarchical clustering was performed using Partek bioinformatics software.

Tissue-resident lung macrophages are derived from embryonic lineages (i.e., primitive yolk sac macrophages or fetal liver monocytes recruited to the lung during embryonic development) or recruited from bone marrow-derived circulating monocytes (i.e., monocyte-derived macrophages) (27–29). Under homeostatic conditions, these tissue-resident macrophages are maintained through self-renewal for the animal’s lifespan or replaced by bone marrow-derived circulating monocytes (29–31). The degree and speed of embryonic-derived cell substitution by bone marrow-derived cells may depend on the organ and cellular location (29, 32–35). To address the ontogeny and maintenance of resident pulmonary macrophages, Cx3cr1^CreER/tdTomato mice were treated with tamoxifen in utero to label primitive yolk sac progenies (E7.5) vs. both yolk sac- and fetal liver monocyte-derived progenies (E13.5). Lineage-labeled mice were allowed to age to adulthood to define the ontogeny of macrophages in adult mice ( Figure 1A ). With E7.5 labeling, approximately 40% of interstitial macrophages (IMØs, CD45⁺CD64⁺CD11b⁺SiglecF^-) were tdTomato⁺ ( Figure 1B ), but 100% of the alveolar macrophage population (AMØ; CD45⁺CD64⁺CD11b^-SiglecF⁺) were tdTomato^- ( Figure 1B ). At E13.5, 60% of IMØs and ~75–80% of AMØs were tdTomato⁺ ( Figure 1B ), suggesting that IMØs and AMØs have different ontogeny, and a substantial percentage of IMØs are derived from primitive yolk sac progenitors.

In most organs, yolk sac-derived tissue-resident macrophages are diluted or replaced by bone marrow-derived macrophages as the animals mature and age (36). To determine the persistence of yolk sac macrophage-derived IMØs and account for labeling efficiency, the percentage of E7.5 lineage-labeled tdTomato⁺ IMØs were normalized to lineage-labeled microglia, as microglia are derived from yolk sac macrophages and persist through the lifespan of animals (37). The ratio of lineage labeled IMØs to microglia remained relatively stable at all ages of animals examined (7, 16, and 32 weeks of age) ( Figure 1C -Age). There were no alterations in the ratio based on the dosage of tamoxifen or the gender of the animal ( Figure 1C -Dosage and Gender). A minimal increase in tdTomato⁺ IMØ proportion was observed based on tamoxifen-induction timing at E7.5 vs. E13.5 ( Figure 1C -Induction), which might be due to improved labeling efficiency of IMØs at E7.5 or a minor contribution of the fetal liver monocyte precursors to the IMØ pool. The lineage labeled IMØs to microglia ratio remained relatively stable with different tamoxifen doses, animal gender, and age ( Figure 1C -Dosage, Gender, and Age). These findings suggest that, under homeostatic conditions, a substantial portion of tissue-resident pulmonary IMØs, like microglia, are derived from yolk-sac progenitors and maintained through local self-renewal. To further confirm the ontogeny of resident pulmonary macrophages, the transcriptome of sorted AMØs, IMØs, microglia, yolk sac macrophages, Ly6C⁺ classical monocytes (Ly6C⁺ Mono), and Ly6C^- non-classical monocytes (Ly6C^- Mono) were examined by microarray analysis followed by hierarchical clustering ( Figure 1D ). The AMØs differed from primitive yolk sac progenitors and clustered with monocytes (Ly6C⁺ and Ly6C^-). The IMØs were distinct from AMØs but shared transcriptomic signatures with primitive yolk sac progenitors. Conflicting reports exist regarding the ontogeny of embryonic-derived IMØs (yolk sac progenitors vs. fetal liver monocytes) (31, 38). However, our findings support that IMØs and AMØs have distinct ontogeny and genetic programming, where AMØs primarily derive from fetal liver monocytes, and a substantial population of IMØs derive from primitive yolk sac progenitors. Additionally, IMØs derived from embryonic progenitors persist through adulthood under homeostatic conditions.

When the niche of tissue-resident macrophages is disturbed through experimental depletion or inflammation, bone marrow-derived macrophages adopt a tissue-resident phenotype with self-renewal capabilities (8, 32, 33, 39–41). After restoring homeostasis following environmental perturbation, bone marrow-derived macrophages often closely resemble those of unperturbed tissue-resident macrophages (8, 41). However, it remains uncertain whether altering the ontogenetic composition of tissue-resident macrophages affects the immune response to subsequent insults. To investigate this, we induced the replacement of embryonic-derived tissue-resident macrophages with bone marrow-derived ones. After homeostasis was re-established, we evaluated the effects of this switch from embryonic to bone marrow-derived macrophages on allergic airway responses to HDM ( Figure 2A ). Macrophage ontogenetic alteration was induced by intranasal exposure to LPS every other week for three doses. The mice were allowed to recover for > 8 weeks to ensure complete resolution of LPS-induced acute inflammation. To assess the efficacy and persistence of replacement at more than 8 weeks after LPS exposure, we examined the change in the proportion of tdTomato⁺ AMØs and IMØs in E13.5 embryonically labeled Cx3cr1CreER/tdTomato mice. Compared to PBS control mice, LPS-exposed mice showed approximately a 50% reduction in lineage-labeled tdTomato⁺ AMØs or IMØs, indicating the recruitment and persistence of bone marrow-derived macrophages ( Figure 2B ).

We assessed whether prior LPS exposure and associated macrophage alteration impacted allergic responses to the HDM challenge. While there were no significant differences in airway responses to methacholine challenges between PBS or LPS pre-exposed animals but not sensitized to HDM (a.k.a. PBS_PBS or LPS_PBS), we observed that mice pre-exposed to LPS and then HDM (LPS_HDM, red open triangle), when compared to mice exposed to PBS and then HDM (PBS_HDM, red open box), had reduced airway physiologic responsiveness, including resistance and elastance ( Figure 2C ). While LPS pre-exposure did not impact the degree of total inflammatory cell infiltrate (H&E) or peri-bronchial fibrosis (Mean % Trichrome), replacement by bone marrow-derived macrophages in the setting of repeated LPS exposure is associated with reduced mucus accumulation in response of HDM exposure (Mean PAS score) ( Figure 2D ; Supplementary Figure S3 ). Additionally, while LPS pre-exposure did not impact overall HDM-induced inflammatory cell infiltration assessed by BALF total cell, macrophage, neutrophil, lymphocyte counts, and proportion of immune cells in lung tissues, there was a significant reduction in eosinophil numbers and proportion in BALF and lung tissues in LPS_HDM animals ( Figure 2E ; Supplementary Figures S4A, B ). Consistent with decreased eosinophils, there was a reduction in T helper 2 (Th2) cytokines in BALF, including IL-4, IL-5, and IL-13, which are known to modulate eosinophil trafficking and functions ( Figure 2F ). The absence of significant differences between PBS and LPS pre-exposed mice without HDM challenge by histologic scoring of inflammation ( Figure 2D ; Supplementary Figure S3 ), immune cell counts in bronchoalveolar lavage fluid (BALF) ( Figure 2E , Supplementary Figure S4A ), immune cell proportions and macrophage counts in lung tissues by flow cytometry ( Figure 2E , Supplementary Figures S4B, C ), and cytokine analyses of BALF (i.e., IL-1β, IL-4, IL-13, IL-1β, IL-5, IL-6, IL-10, IL-13, TNFα, and KC)( Figure 2F ) are consistent with the absence of persistent inflammation resulting from prior LPS-exposure before subsequent HDM challenge. These data support that prior LPS exposure, which alters the ontogenetic composition of AMØs and IMØs, reduces HDM-induced mucin production, Th2 cytokines, and eosinophil infiltration, and improves allergic airway responses to HDM.

To determine the impact of LPS pre-exposure and macrophage ontological alterations on HDM allergic responses, we performed single-cell RNA-sequencing (ScRNAseq) to define immune cell composition and gene expression. To define pulmonary macrophage ontogeny, we generated bone marrow chimera by transferring CD45.1 bone marrow donor cells into busulfan-treated CD45.2, embryonic-labeled (at E13.5) Cx3cr1^CreER/tdTomato recipient mice ( Figure 3A ). We have previously shown that busulfan treatment induces efficient bone marrow myeloid ablation while having minimal effects on tissue-resident pulmonary macrophages (8, 11). Thus, we can clearly distinguish embryonic-derived resident macrophages (CD45.2⁺) from those arising from the bone marrow compartment (CD45.1⁺). After engraftment, consistent with preservation of tissue-resident pulmonary macrophages following busulfan treatment, >80% of AMØs and approximately 40% of IMØs were identified as CD45.2+ (host origin) ( Supplementary Figure S4D ). Chimeric mice were exposed to PBS or LPS every other week for 3 doses, allowed to recover for >8 weeks, and exposed to PBS or HDM. After LPS exposure, approximately 25% of AMØs and only 5–10% of IMØs were CD45.2+ after repeated LPS exposure consistent with altered ontogenetic composition following LPS exposure ( Supplementary Figure S4D ).

For ScRNAseq, myeloid cells were enriched, and embryonic- and bone marrow-derived cells were segregated by flow sorting for live, CD45⁺ Ly6G^- B220^- cells and then separated by CD45.1 vs. CD45.2 ( Figure 3A ). ScRNAseq of sorted cells identified distinct populations of cells based correlation profiles with bulk RNA-seq from the Immunological Genome Project (ImmGen) database (25) and confirm by cell type-associated markers. Clustering identified expected clusters of macrophages (Trem2 and Cxcl2), T cells (Cd3e and Cd3g), dendritic cells (Xcr1 and Fscn1), and monocytes (Cx3cr1) ( Figure 3B ). Small clusters of residual endothelial cells, B cells, and epithelial cells were observed which were excluded from subsequent analyses ( Figure 3B ). To define if there are unique clusters based on exposure condition and ontogeny, macrophages were subclustered into AMØs (Itgax and Car4) and IMØs (Itgam, Csf1r, Cx3cr1, Ccr2, and Lyve1) groupings ( Figure 3C ) as well as subclustered based on both exposure conditions (PBS vs. LPS, and PBS vs. HDM) and ontogeny (CD45.1 vs. CD45.2)( Figure 3D ). Even though there were no significant pathway differences by IPA analyses (defined as Z score >2 or <2) between IMØs from the PBS or LPS pre-exposed animals without subsequent HDM challenge, following HDM exposure, we observed an expansion of IMØs in a specific cluster of cells ( Figure 3D , dark blue) arising primarily from CD45.2 embryonic-derived interstitial macrophages in the setting of the HDM challenge. These observations suggest the unique macrophage clustering based on ontogeny and exposure conditions.

AMs and IMs have been implicated in regulating allergic asthma responses; thus, we performed further analyses of both AMØs and IMØs to determine how altering the ontogenetic milieu affects their transcriptomic profiles. Sub-cluster analysis of AMs revealed 5 unique clusters (AM1–5; Figures 4A, B ). Based on differential gene expression and Ingenuity pathway (IPA) analysis ( Supplementary Figures S5 , S6A ), AM1 consisted primarily of cells that down-regulated pro-inflammatory and proliferative processes (e.g., interferon signaling, chemokine signaling, neuroinflammatory signaling, mitosis, metaphase signaling, and cell cycle control of chromosome replication); AM2 represents a cluster of proliferating cells (e.g., Mki67; DNA replication, mitosis, and cell cycle progression); and AM3–5 participated in immune responses (e.g., myeloid cell activation, chemotaxis/migration/accumulation, phagocytosis, or systemic autoimmune syndrome). Overlaying clusters based on exposure conditions and ontogeny did not reveal unique condition/ontogeny-specific clustering by UMAP or the individual cell/cluster percentages ( Figure 4C ). No prominent contribution of macrophage ontogeny (CD45.2 vs. CD45.1) was observed to HDM responses; however, DEG-based clustering revealed differences between the pre-HDM exposure conditions (PBS vs. LPS pre-exposure) ( Supplementary Figures S5A–E ). This data suggests that LPS pre-exposure, but not ontogeny, modulates AMØs phenotype in response to HDM challenge.

Analyses of IMØs identified 7 distinct clusters ( Figure 5A ). Similar to AMØ clusters, differential gene expression and IPA analyses define cells that down-regulated proliferative pathways (IM1), upregulated proliferation pathways (IM5), and cells involved in overlapping inflammatory pathways (IM3, 4, 6, and 7; e.g., iNOS, IL-1, IL-6, and Trem-1 pathways) ( Figure 5A ; Supplementary Figure S6B , and data not shown). Interestingly, IM2 consisted mainly of the population of previously observed cells (dark blue) that arise primarily from CD45.2 embryonic-derived Interstitial macrophages following the HDM challenge (CD45.2 PBS_HDM, blue in Figures 3D , 5B, C ). This response was not observed in animals re-challenged with PBS regardless of ontogeny or prior LPS exposure (CD45.1 PBS_PBS, CD45.2 PBS_PBS, CD45.1 LPB_PBS, and CD45.2 LPS_PBS) and animals pre-exposed to LPS followed by HDM challenge regardless of macrophage ontogeny (CD45.1 or CD45.2 LPS_HDM) as there was a minimal cellular contribution to the unique IM2 sub-cluster (dark blue cells) by these samples ( Figure 5C ). Additionally, top gene analysis identified that the IM2 cluster was defined by expression of the Chi3l3 (chitinase-like 3/Ym1), which has been previously associated with Th2, eosinophilic, and allergic airway responses ( Figure 5D ) (42–45). The Chi3l3 gene expression level was also most prominent in the CD45.2 PBS_HDM within the IM2 subcluster (Chi3l3 ^hi; dark blue cells). These findings suggest that Chi3l3 ^hi CD45.2 IM2 cells uniquely and dynamically increased following HDM exposure and could potentially regulate allergic airway responses. Therefore, as opposed to AMØs, replacing embryonic-derived interstitial macrophages with bone marrow-derived ones is associated with dampening of HDM-induced allergic inflammation.

To better define the genetic pathways defining resident macrophages in the IM2 cluster, we performed an analysis based on ontogeny and exposure. A heatmap of top differentially expressed genes based on either the PBS_HDM or the LPS_HDM exposures and lineage (CD45.1 vs. CD45.2) of the IM2 cluster was generated ( Figure 6A ). This dataset revealed clear separation in the gene expression pattern in the CD45.2 PBS_HDM cells from other conditions (CD45.1 PBS_HDM, and CD45.1 vs. CD45.2 LPS_HDM). To clarify the pathways involved in this response, we performed an IPA pathway analysis to assess the potential functions of CD45.2 PBS_HDM IM2 cells. We focused on genes and pathways related to macrophage functions that could regulate allergic asthma, including processes related to eosinophil signaling, inflammation, vascular-associated signals, remodeling, and metabolism/death. Consistent with high Chi3l3 expression, the CD45.2 PBS_HDM cells in IM2 were enriched for many macrophage-mediated inflammatory and remodeling processes relevant to allergic airway responses. These included those that regulate eosinophilic responses (e.g., CCR3 signaling in eosinophils, Fc Epsilon RI signaling, and IL-3) ( Figure 6B ). These pathways were downregulated in the LPS_HDM cell groups. In addition, we observed that CD45.1 bone marrow-derived macrophages (CD45.1_PBS_HDM) exhibited reduced allergic pathway activation compared to CD45.2 embryonic-derived macrophages (CD45.2_PBS_HDM), suggesting that replacement of CD45.2 embryonic-derived macrophages by CD45.1 bone marrow-derived macrophages reduces macrophage-derived allergic responses. While to a lesser degree than the difference observed across different ontogeny (CD45.2 vs. CD45.1), we also observed a reduction of allergic pathway signaling in CD45.2 embryonic-derived tissue-resident macrophages following LPS exposure (CD45.2_LPS_HDM) when compared to PBS exposed CD45.2 tissue-resident macrophages (CD45.2_PBS_HDM) suggesting a decrease of tissue-resident macrophage allergic potential in addition to replacement by CD45.1 cells ( Figure 6B ). These findings further indicate that there may be effects related to replacement by CD45.1 cells and the impact of prior exposure on the allergic potential of tissue-resident macrophages. We also observe that LPS pre-exposure leads to the downregulation of small numbers of inflammatory pathways (i.e., CCR5, IL-8, and thrombopoietin) in a lineage non-specific fashion ( Supplementary Figure S7 , Teal). As expected, there were upregulated pathways in response to HDM, which were shared across ontogeny or pre-exposure conditions ( Supplementary Figure S7 , Yellow). When taken together, these findings support that the ontogeny of interstitial macrophages is important for subsequent allergic responses to an allergen.

Given the identification of a unique IMØ cluster, which upregulated genes and pathways associated with HDM pulmonary allergic responses, we then attempted to define their location in lung tissue. By assessing the top gene expression of individual clusters based on exposure and ontogeny ( Supplementary Figure S8 ), we identified Mrc1/Cd206 (a macrophage marker) and Ctsk as uniquely co-expressed genes in the IM2 cluster ( Figures 7A, B ). Immunofluorescence analysis of lung tissue sections revealed a unique population of CD206⁺CTSK⁺ cells located in the parenchyma of the terminal bronchioles in the lungs of PBS_HDM treated mice ( Figure 7C ; Supplementary Figure S9 ). Consistent with our observations in ScRNAseq where, in PBS_HDM treated mice, a unique population of CD45.2 cells was evident and enriched for gene and pathways associated with allergic responses, the CD206⁺CTSK⁺ macrophages were only observed in alveolar regions adjacent to terminal bronchioles of PBS_HDM treated animals. Furthermore, CD206⁺CTSK⁺ double-positive cells were not observed in other conditions ( Figure 7C , PBS_PBS, LPS_PBS, or LPS_HDM). This observation supports tissue locational evidence of a unique IMØ population following HDM exposure, central to regulating allergic airway responses.